Multi-category human motion recognition based on MEMS inertial sensing data

This paper presents multi-category human motion recognition methods based on MEMS inertial sensing data. A micro inertial measurement unit (muIMU) that is 56 mm*23 mm*15 mm in size was built. This unit consists of three dimensional MEMS accelerometers, gyroscopes, a Bluetooth module and a MCU (Micro Controller Unit), which can record and transfer inertial data to a computer through serial port wirelessly. Five categories of human motion were recorded including walking, running, going upstairs, fall and standing. Fourier transform was used to extract the feature from the human motion data. The concentrated information was finally used to categorize the human motions through CNN (cascade neural network) SVM (support vector machine) and HMM (hidden Markov model) respectively. Experimental results showed that for the given 5 human motions, HMM have the best classification result with correct recognition rate range from 90%-100%.